Potter T.D., Colman B.R. (co-chief editors). The handbook of weather, climate, and water: dynamics, climate physical meteorology, weather systems, and measurements
Подождите немного. Документ загружается.
must be capable of operating under adverse weather conditions where freezing
rain, ice, and snow can be expected. Potential benefits of this study are very
large in that, for a small expense, a considerable increase in water runoff from
melting snow may occur. This water runoff is then used to generate hydro-
electric power.
8. Long-range transport of pollutants. This study was initially designed to
describe the fate of pollutants from one or more power plants as they were
transported to downwind areas far removed from them. This study was
co-sponsored by several governmental groups and had the mission of devel-
oping models for both air quality and acid deposition over distances of
hundreds of kilometers. Meteorological sensors used in this project included
many in situ and short-path sensors along with long-path sensors, upper air
sensors, and a variety of aircraft meas urements.
7 EXAMPLE OF DATA PROCESSING PROCEDURES
As part of the transmission line loading (dynamic thermal rating) effort, a simple
prearchiving program was developed for in situ wind speed and direction measure-
ments where inexpensive on-site digital recording systems were used. Data resulting
from this program provide basic data sets that may be somewhat applicable to other
project studies. In general the program was designed to collect, generate, store, and,
on demand, transmit the following data for any desired time period:
1. Peak scalar speed
2. Time of peak speed
3. Unit vector wind direction at time of peak speed
4. Mean scalar wind speed over sampling period
5. Mean unit vector wind direction over sampling (period excludes observations
of calm winds)
6. Mean resultant vector wind direction over sampling period
7. Standard deviation of wind speeds over sampling period
8. Standard deviation of unit vector wind directions over sampling period
9. Standard deviation of resultant vector wind directions over samplin g period
10. Standard deviation of nonoverlapping 1-min mean speeds over sampling
period
11. Standard deviation of nonoverlapping 1-min mean unit vector directions over
sampling period
If the desired sampling period was longer than the basic time of 10 min, e.g., 1 h,
then values for items 10 and 11 would be repeated for longer term means of 2, 5, or
even 10 min. The purpose of generating the many standard deviations from averaged
values was to detect the energy-containing frequencies within the lengths of record
846 SURFACE LAYER IN SITU OR SHORT-PATH MEASUREMENTS
considered. More extensive or different analyses of incoming raw data can be easily
programmed, but cost–benefit relationships as well as storage capacities and data QC
must be consid ered. Again the example is intended only to be an illustration of a
practical data processing approach and not a recommended guideline.
8 CONCLUSIONS
The above discussion of the recent history of meteorological support for an electric
utility and of programs used within a selected utility are not complete but do serve as
an indication of the importance of high-quality measurement programs as well as the
need to match measurement requirements with study efforts. Hopefully, it also
provides some insight into the need for making as complete as pract ical measure-
ments for any given study so resulting data packages can be used in other studies.
Other electric utilities in different areas of the country, or world, will undoubtedly
have a different list of meteorologically related problems or concerns such as
lightning protection and cooling water temperatures.
REFERENCES
ANS (2000a). American Nuclear Society, Determining Meteorological Information at Nuclear
Facilities, La Grange Park, IL.
ANS (2000b). ANS=ANS-3.11-2000, American Nuclear Society, La Grange Park, IL.
U.S. EPA (1989). Quality Assurance Handbook for Air Pollution Measurement Systems. Vol. IV,
Meteorological Measurements, prepared by Thomas J. Lockhart for the U.S. Environmental
Protection Agency, Research Triangle Park, NC.
REFERENCES 847
CHAPTER 44
INDEPENDENT AUDITING ASPECTS
OF MEASUREMENT PROGRAMS
ROBERT A. BAXTER
There are a number of aspects that need consideration in the design and execution of
measurement programs to assure the data collected are of documented quality and
meet the program data quality goals. This chapter looks at the independent auditing
aspects of monitoring as an integral tool to the overall data collection effort and
provides examples of how independent audits contribute to the understandi ng of the
quality of the data collected.
Prior to entering the role of audits, it is helpful to understand the details of
monitoring programs and see where audits fit into the overall data collection
scheme. The monitoring program and its goals are described in a monitoring plan.
1 MONITORING PLAN
A monitoring plan is a general description of the overall plan to collect data. A
monitoring plan includes the monitoring program goals, methods of data collection,
locations where data will be collected, internal and external checks of the measure-
ment program, and the overall management and reporting structure. It is typically
prepared well in advance of the collection of data so that management and=or
regulatory review can identify and resolve any deficiencies in the measurement
program.
Once the monitoring plan is agreed upon and ready to be implemented, a quality
assurance project plan is prepared that addresses all of the details of the data
collection program.
Handbook of Weather, Climate, and Water: Dynamics, Climate, Physical Meteorology, Weather Systems,
and Measurements, Edited by Thomas D. Potter and Bradley R. Colman.
ISBN 0-471-21490-6 # 2003 John Wiley & Sons, Inc.
849
2 QUALITY ASSURANCE PROJECT PLAN (QAPP)
A quality assurance project plan is a formal docum ent describing in comprehensive
detail the necessary activities that must be implemented to ensure that the results of
the work performed will satisfy the stated performance criteria.
The QAPP defines all details of the program including the methods of data
collection, the specific instrumentation used for collection of each variable, methods
of calibration, data storage, backup, validation, and reporting. Also included is the
audit plan to verify the implementation of the methods and procedures defined in
the QAPP. The U.S. Environmental Protection Agency (EPA) provides guidance for
the preparation of QAPPs for environmental monitoring programs (U.S. EPA, 1998).
After reading the guidance and recognizing the recommended level of detail in a
QAPP, one may ask the question ‘‘Are all of the checks and documentation really
necessary? After all, I have been collecting data using a variety of measurement
methods for many years.’’ The answer is a qualified yes. The qualification steps back
to what the data will be used for and how defendable one wants to make it. Backyard
weather forecasting requires little in oversight and documentation. Permit applica-
tions and compliance on the other hand requires well-documented, defensible data.
One learns through years of experience that even the most competent professional
or scientist can make mistakes through the redundant process of setup, data collec-
tion, and data validation. These mistakes may be minor but could have a detrimental
effect on the quality of the collected data. In some cases, quality and the proper way
to operate a system may be compromised because of budget limitations or lack of
physical resources. It is extremely important to identify the instances when data has
been compromised and then quantify the impact it has on the quality of the data
reported. This will provide end users with records and information needed to assess
whether the collected data will meet their needs.
A properly designed program will have cross checks built into the overall
measurement plan. These checks will be defined in the QAPP in the sections on
quality control and quality assurance.
Quality Control (QC) The overall system of technical activities that measures
the attributes and performance of a process, item, or service against defined
standards to verify that they meet the stated requirements established by the client.
Quality Assuran ce (QA) An integrated system of management activities
involving planning, implementation, assessment, reporting, and quality improve-
ment to ensure that a process, item, or service is of the type and quality needed and
expected by the client.
Quality control activities are designed to control the quality of a product so that it
meets the user’s needs. This includes the routine calibrations, data validation, preven-
tive maintenance, equipment certification, etc. Quality assurance is the process
whereby the implementation of the QC program and other activities is checked
and verified. It encompasses the various activities needed to assure the QC program
850 INDEPENDENT AUDITING ASPECTS OF MEASUREMENT PROGRAMS
is being implemented and is working. QA includes QC as one of the activities
needed to ensure that the product meets defined standards of quality. Details on
quality assurance in meteorological measurements can be found in U.S. EPA (1995).
A key element in QA is the use of independent audits to review operations and
make reports to manag ement on the status of the measurement program. These
audits are performed by an individual or group that is independent of those
making the measurements. This independence allows the identification of potential
problems without any conflict of interest the findings may have with either the
technical merit of the data or financial resources needed to correct deficiencies.
Audit A systematic and independent examination to determine whether quality
activities and related results comply with planned arrangements and whether these
arrangements are implemented effectively and are suitable to achieve objectives.
In performing the audits, there are specific roles of the audito r and auditee.
Auditor A person that is qualified to perform audits.
Auditee Person or organization that is operating a measurement program and is
the one being audited.
Ideally, the auditor will be an expert in the field of the measurements being
performed. While not always true, the primary prerequisite is that the auditor under-
stand the measurements being performed and be able to identify if the methods
followed are consistent with the monitoring and quality assurance plans. The auditor
is always a guest at the meas urement site and should not perform any of the instru-
ment removal or other activities that are part of the normal operation of the site. In
past instances the auditor has helped in the removal of equipm ent and performance
of duties for the operator and inevitably an instrument breaks or becomes inoperable
that becomes attributable to the audit or.
3 CASE STUDIES
Given the above overview of the elements in the planning and oversight of a
measurement program, it is useful to review some case studies to demonstrate the
value in independent audits to support overall data collection and documentation.
The case studies below descri be field experiences in auditing some aspects of
meteorological measurement programs.
In this first case study, a major measurement program was carried out in the
western United States that included measurements made by federal, state, and
local agencies as well as private contractors. The design of the program included
an independent QA contractor responsible for system and performance auditing of
the installation and operation of the surface and upper air meteorological systems.
The surface meas urements included wind measurements using conventional cup-
3 CASE STUDIES 851
and-vane or propeller-vane anemometers. Some of the systems were part of existing
measurement programs while the majority were installed specifically for this study.
The auditing plan for the study called for initial siting audits that assessed the
appropriateness of the selected sites for the measurements. The results of these audits
helped program management determine if changes were needed in the selected sites.
Once final sites were selected and instruments installed, audits were performed on
the surface meteorological systems. While there may have been several specific
problems noted for each of the sites, there was a common equipment alignment
problem noted with one of the measurement groups. This problem was present at
most of their sites. The alignment and orientation of sensors have always presented
challenges. While the most common method of orientation is to use the local
magnetic declination to cor rect the alignment to true north, local anomalies in the
magnetic field can create errors of 10
or more. With this par ticular measurements
group (called group A), the number of sit es set up and operated led them to stream-
line the alignment process using a hand-held ‘‘data scope.’’ The scope allowed the
direct entry of the local declination to the magnetic readings making the alignment
checks quick and easy. A second group (called group B) used magnetic methods to
determine the alignment but did not apply a specific declination. Instead, the
magnetic readings were corrected to true north based on a measurement of the
sun’s azimuth angle and the calculated true angle of the sun for the site’s latitude
and longitude.
Figure 1 shows the alignment audit results from groups A and B for the surface
sensors. In the overall program plan, the data quality objective for the wind direction
Figure 1 Surface wind direction sensor alignment results from two different measurement
groups.
852
INDEPENDENT AUDITING ASPECTS OF MEASUREMENT PROGRAMS
data was established as 5
, as indicated in the figure. The results of the alignment
audits showed group A had much more scatter in the alignment accuracy, which was
a direct result of the method used for alignment. The problem was systematic
throughout the sites set up by g roup A. Agreement was reached between the auditee
and auditor on appropriate methods for alignment of the systems. The procedures
used by the auditee were modified to obtain the nee ded accuracy. The audit in this
case served as a valuable training and teaching exercise for group A performing the
measurements and corrected a long-term systematic problem in aligning sensors.
In a second example, a station was audited that collected both air quality and
meteorological data. Meteorological sensors were located on a 10-m tower adjacent
to the trailer. This particular station demonstrated an all-too-familiar example of how
the meteorological sensors generally take second place to the air quality measure-
ments. Due to the relative complexity of the air quality instr umentation, much of the
maintenance effort focused on those instruments. The meteorological sensors were
set up, operation verified, and then the sensors left alone to collect data without any
further checks. This is a common scenario that arises from the ability to visually look
at the sensors, see them rotating in the wind and aiming in appropriate directions.
This gives the impression that the equipment is operating acceptably and no further
checks need to be performed. To the contrary, bearings in the wind sensors may fail,
cups or propellers may be damaged, and potentiometers that convert the vane
direction to an electrical signal may wear. Other sensors such as those used for
temperature may become corroded and produce erroneous signals. The audit in
this example looked at wind and temperature measurements and included a
system and siting audit to determine the appropriateness of the site for the intended
measurements.
Results of the audit showed the following:
1. The entire meteorological system had not been serviced since installation with
no calibrations performed.
2. The temperature sensor had corroded into the radiation shield and could not be
removed to verify its proper calibration.
3. The wind direction sensor had corroded into the mounting and the connector
failed when removed from the sensor. This hampered the performance
evaluation of the sensor.
4. The wind speed bearings had corroded to the point where the starting
threshold of the sensor was almost 2 m=s.
5. The alignment of the wind direction sensor was incorrect, resulting in a fixed
offset in the measurements of nearly 10
.
While the items noted above are critical in the collection of valid data, the biggest
problem was in the siting of the sensors. Figure 2 shows the locat ion of the measure-
ments relative to a nearby building and pollution sources. The building was about
8 to 10 m high and located about 20 m from the sensors. The problem was
compounded by the proximity to the major air pollution source that was the focus
3 CASE STUDIES 853
of the overall measurement program. Located to the south was an industrial park
with many sources of pollution. This air quality and meteorological station was
intended to document the transport of pollutants from the industrial park to the
station, which was located adjacent to residential housing. The data was then used
in subsequent computer modeling of the sources to determine the impacts. With the
building between the station and the source, winds from the south would be signifi-
cantly altered and not be representat ive of the area meteorological conditions. The
results of the indepe ndent audit noted these problems and made recommendations
on where an acceptable site could be located and how the sensor s should be main-
tained to collect accurate, defendable data. The five sensor-specific items noted
Figure 2 Location of meteorological sensors relative to an adjacent building and nearby
pollution sources. See ftp site for color image.
854
INDEPENDENT AUDITING ASPECTS OF MEASUREMENT PROGRAMS
above were highlighted and recommended servicing intervals provided in accord-
ance with EPA guidance. The results of the audit will eventually lead to data that can
be used for the intended modeling purposes.
The third case study draws more on the role of quality auditing in the planning
and execution of a measurement program in the early stages of setup. As part of a
large air quality and meteorological measurement program, the quality assurance
contractor held a workshop for all who were making meteorological measurem ents
as well as those who were performing the audits. The purpose of the workshop was to
introduce the QA program to all measurement personnel and identify the needs and
requirements of the program. This included the audit procedures, criteria used for
passing or failing the audits, and the overall data quality objectives. In this manner
all personnel making the measurements would understand the common goals of the
program and be prepared for the a uditing steps that would follow.
All participants in the program had many years of experience in the measure-
ments being performed. One in particular had indicated objections to the workshop
and the time it would take to go through the ‘‘orientation’’ process when it still had
half of its 35 plus sites to install. This contractor had vast experience in setup of
large networks of meteorological instrumentation on the east coast of the United
States. Reluctantly, but fortunately, they agreed to participate. As part of the work-
shop exercise, one of the tasks was for all auditors and measurement contractors to
measure the true direction of a distant landmark. A variety of methods were
employed by the participants and the results were compa red. For reference, the
workshop director, who was also the technical director for the meteorological
audit program, used his measurement as the ‘‘standard’’ by which all others were
compared. For the most part all results were within 2
. One participant was heard
mumbling and groaning, indicating his answer differed by more than 30
. After a
brief review of the method used to determine the true pointing direction, it was
learned that he had applied the local declination in the wrong direction. This
happened to be the contractor with extensive east coast experience where the local
declination is a pplied in the opposite direction. It was also the same contractor that
had responsibility for setup of over 35 stations, half of which were already in
operation with the improper alignment, an error that was eventually corrected.
The lesson of this story is that the independent audit program implemented at the
start of the field study identified and corrected a major problem early. This allowed
all participating in the project to understand the requirements, perform the measure-
ments in a consistent manner, learn from each other’s expertise, and ultimately
collect data of known quality that are defensible.
As a final example, it is valuable to address the traditional challenge of calcu-
lating an ‘‘average’’ wind direction. This is not a trivial problem since the wind
direction is a circular function that is not amenable to simple analog averages. For
years there have been debates on the appropriate method to handle the average,
whether it be dealt with on a vector basis or through some algorithms that deal
with the north crossing through 360
. The EPA has provided guidance on procedures
for calculating wind direction in regulatory driven and other monitoring programs
(U.S. EPA, 2000) that have been incorporated in one form or another into commer-
3 CASE STUDIES 855